EP3112854A1 - Method and device for the thermomechanical analysis of a sample - Google Patents

Abstract

The invention relates to a method for the thermomechanical analysis of a sample (P) of a material, comprising (a) arranging the sample (P) in a thermomechanical analysis device (10) and tempering the sample (P) by means of the analysis device according to a modulated temperature program (b ) Recording data obtained by means of the analyzer which are representative of a change in length (dL) of the sample (P) in the course of tempering, (c) evaluating the data to obtain a reversible component (dL rev) of the change in length (dL) and / or a reversible component (± rev) of the thermal expansion coefficient (±) of the sample (P), (d) calculating a corrected reversible component (dL rev-corr; ± rev-corr) of the change in length (dL) and / or coefficient of thermal expansion (±) by means of a correction parameter (k) obtained by using a ratio of a total change in length (dL total) determined from the data arakterisierenden parameter (± total) and a reversible component (dL rev) of the change in length (dL) characterizing parameter (± rev) is calculated. According to the invention, the temperature program has a first segment (S1) with a first base heating rate (²1) and the correction parameter (k) becomes temperature-dependent on data from the first segment (S1) from an estimate of a temperature-dependent function of said ratio in a region of the first segment (S1) is calculated, in which the sample (P) is not subject to a thermally induced conversion process, wherein the temperature program, a second segment (S2) with a z. B. relative to the first Basisheizrate (²1) smaller second base heating rate (²2) and for a calculation of the corrected reversible component (dL rev-corr; ± rev-corr) of the change in length (dL) and / or the coefficient of thermal expansion (±) im second segment (S2) of the calculated using the data from the first segment (S1) correction parameter (k) is used.

Description

The present invention relates to a method for thermomechanical analysis of a sample according to the preamble of claim 1 and to a device for thermo-mechanical analysis according to the preamble of claim 12.

Thermomechanical analysis, often referred to as TMA ("thermomechanical analysis"), measures one or more mechanical properties of a sample of a material as a function of temperature.

The material may be z. B. act solid, liquid or pasty material. Often this is z. B. a change in length as a function of temperature and / or time under well-defined mechanical load of the sample is measured, the sample is usually a predetermined time-dependent temperature, ie a "temperature program" is exposed.

In particular, in the case of a practically negligible (relatively small), z. B. only by the use of a Fühlstempels or the like for measuring the change in length-related mechanical stress of the sample, the TMA is often referred to as dilatometry, or when using a modulated temperature program as a temperature-modulated dilatometry.

In a TMA of the type of interest here, a "modulated temperature program" is used, which means that the time course of the temperature is composed of an underlying, mostly linearly predetermined temperature change corresponding to a "base heating rate", and from a temperature change overlying, usually sinusoidal ( alternatively conceivable, for example, triangular, rectangular or sawtooth-shaped) predetermined temperature modulation.

An important goal of a TMA of the type of interest here is, for the material of the sample, at least one "reversible component" of the (change in temperature-related) change in length and / or in particular z. B. to determine a reversible component of the thermal expansion coefficient.

Since, if the sample is subjected to a thermally induced conversion process (eg phase transition, glass transition, sintering process, etc.) during thermomechanical analysis, an immediately measured "total component" of the change in length or coefficient of thermal expansion is composed of a reversible component "and a" non-reversible component ", after determining the" reversible component ", it is also possible simply to determine the" non-reversible component "of the change in length or of the thermal expansion coefficient (by subtracting the reversible component from the total component).

The temperature-dependent determination and characterization of shrinkage or expansion effects of the sample or of the influence of such effects on the change in length or the coefficient of thermal expansion is often a "main goal" of TMA in practice.

1. a) Anordnen der Probe in einer thermomechanischen Analyseeinrichtung und Temperieren der Probe mittels der Analyseeinrichtung gemäß eines modulierten Temperaturprogrammes,
2. b) Aufzeichnen von mittels der Analyseeinrichtung gewonnener Daten, die repräsentativ für eine Längenänderung der Probe im Verlauf der Temperierung sind,
3. c) Auswerten der Daten, um eine reversible Komponente der Längenänderung und/oder eine reversible Komponente des thermischen Ausdehnungskoeffizienten der Probe zu ermitteln,
4. d) Berechnen einer korrigierten reversiblen Komponente der Längenänderung mittels eines Korrekturparameters, der als das Verhältnis einer totalen Längenänderung und einer reversiblen Komponente der Längenänderung berechnet wird.

A generic method for thermo-mechanical analysis, for example, in US 6,007,240 described and includes the following steps:

1. a) arranging the sample in a thermomechanical analysis device and tempering the sample by means of the analysis device according to a modulated temperature program,
2. b) recording data obtained by means of the analysis device, which are representative of a change in length of the sample during the course of the temperature control,
3. c) evaluating the data to determine a reversible component of the change in length and / or a reversible component of the thermal expansion coefficient of the sample,
4. d) calculating a corrected reversible component of the change in length using a correction parameter calculated as the ratio of a total change in length and a reversible component of the change in length.

The Fig. 1 to 3 exemplify a composition of a linear temperature change ( Fig. 1 ) and a sinusoidal temperature modulation ( Fig. 2 ) to Realization of a modulated temperature program ( Fig. 3 ). In the Fig. 1 to 3 in each case the time t is plotted to the right and the temperature T is plotted upwards.

Fig. 1 shows a linear change in temperature T, which can be described as follows: T = T ₀ + (β × t). Here, T ₀ denotes an initial temperature and β the heating rate. In the example of Fig. 1 the heating rate is β 2.0 K / min. It should be noted that here and also in the description below the invention for the sake of simplicity is always spoken of a heating rate β, although this heating rate β can also be specified negative (ie a "cooling rate" accordingly).

Fig. 2 shows a sinusoidal change in temperature T, which can be described as follows: T = T _avg + A _T × sin (ωt). In this case, T _{avg denote} a temperature (over the period _duration ) averaged, A _T an amplitude of the temperature modulation, and ω a modulation frequency. In the example of Fig. 2 T _{avg is} about -1.2 ° C, A _{T is} 1.5 K, and ω is about 2.1 min ^-1 (corresponding to a period of 3.0 min).

Fig. 3 shows the modulated temperature program corresponding to a superposition of the linear change ( Fig. 1 ) and the sinusoidal change ( Fig. 2 ), which can be described as follows: T = T ₀ + (β × t) + A × _T × sin (ωt).

In a TMA carried out with it all parameters of the modulated temperature program (here: To, β, A _T and ω) z. B. the specific application adapted by a user defined (default).

Even with a modulated temperature program, it is _{possible to} specify a temperature T _avg (over the period _duration ). For these, T _avg = T ₀ + (β × t). Thus, the modulated temperature program can also be described as follows: T = T _avg + A _T * sin (.omega.t).

If we define to describe the modulation-induced 'temperature oscillations "a temperature deviation AT = A _T × sin (.omega.t), then the modulated temperature program can finally also be described as follows: T = T _avg + .DELTA.T.

The "thermal expansion coefficient α" is an important parameter that describes the behavior of a material with respect to changes in its dimensions in the case of temperature changes. The thermal expansion coefficient α (only dependent on the temperature) is a material-specific material constant, which is usually more or less self-temperature dependent. The coefficient of thermal expansion α is here understood in terms of a coefficient of linear expansion (in contrast to the thermal expansion coefficient γ often used), and is the proportionality constant between the temperature change dT and the (reversible) relative change in length dUL of the material: dL / L = α × dT. It therefore applies: $\alpha =\frac{1}{L}\times \frac{\mathit{dL}}{\mathit{dT}}\mathrm{,}$

In the situation according to Fig. 1 (linear temperature change) dL = L × α × dT and dT = β × dt and thus: dL = L × α × β × dt.

From this the thermal expansion coefficient α can be determined as follows: $$\alpha =\frac{1}{\mathit{L\beta }}\times \frac{dL}{dt}$$

The in Fig. 1 The linear progression of the change in length dL / L ₀ as a function of the time t is equivalent to the fact that α in this (idealized) example is a temperature-independent constant.

In the situation according to Fig. 3 (Modulated temperature program with linear underlying heating) there are two components in the change in length, namely one (due to the base heating rate β ≠ 0) "behind component" and one (due to the temperature oscillations) "oscillating component". An "underlying dL" can be calculated as an average over a period and is referred to below as dL _total . Here dL _total = L _total × α × dT _avg and dT _avg = β × dt, and thus: dL _total = L _total × a × β × dt.

From this, the thermal expansion coefficient α _total can be determined from the dL _total as follows: $${\alpha }_{\mathit{total}}=\frac{1}{{\mathit{L}}_{\mathit{total}}\mathit{\beta }}\times \frac{d{L}_{\mathit{total}}}{dt}$$

The in Fig. 3 linear course of the change in length dL _total / L ₀ as a function of the time t is equivalent to the fact that α _total in this (idealized) example is a temperature-independent constant.

In the situation according to Fig. 2 (Temperature modulation) is based on the assumption that "T" is the uniform temperature of the respective material sample, also ΔL = L × α × ΔT, and ΔT = A × _T × sin (ωt), resulting in a "change in length" ΔL: $$\mathrm{.DELTA.L}=\mathrm{L}\times \mathrm{\alpha }\times {\mathrm{A}}_{\mathrm{T}}\times \sin \left(\mathrm{.omega.t}\right)\mathrm{,}$$

This is correct, however, only for an "ideal" situation in which the sample temperature in each point of the sample is the same size, ie in particular z. In the center and on the surface of the sample. However, this is not the case in practice since it would presuppose that the modulation frequency ω would be infinitely small or the temperature conductivity of the sample material infinitely large.

Defining an amplitude of the change in length as A _L to describe the modulation-related "length oscillations", the following applies: $$\mathrm{.DELTA.L}={\mathrm{A}}_{\mathrm{L}}\times \sin \left(\mathrm{.omega.t}\right)\mathrm{,}$$

A _L is the amplitude of the measured modulated changes in length. It can be made of a measuring signal representative of these changes in length z. B. by a Fourier analysis (eg., "Fast Fourier Transformation") are calculated.

From this, the (reversible) thermal expansion coefficient α _rev from the oscillating component of the measurement signal can theoretically be determined for ideal situation as follows: $${\mathrm{\alpha }}_{\mathrm{rev}}=\frac{1}{L}\times \frac{A_L}{A_T}$$

In practice, however, this also results in the example of Fig. 2 expressed problem that the temperatures in different areas of the sample more or less differ from each other and from the temperature profile, which is determined by a temperature control device used in the thermomechanical analyzer used according to the modulated temperature program. Due to the external heating and cooling of the sample, in practice a more or less large, not to be neglected temperature gradient always arises in the interior of the sample.

So comes in practice due to the not infinitely rapid heat conduction of heat into the sample and out of the sample out z. B. to a "lagging" of the sample temperature, in Fig. 2 leads to a temporal shift of the length oscillations (curve dL) relative to the temperature oscillations (curve T). At the same time, this effect to cause the average value (averaged over all of the sample) the temperature _T is smaller than the amplitude A specified by the temperature program fails. This in turn has the consequence that the length change amplitude A _{L is} smaller.

In reality, the temperature amplitude in the center of the sample is less than the predetermined temperature amplitude A _T at the sample surface (eg, by a user setting). Therefore, the amplitude of the length oscillations of the sample A _{L is} smaller than in the mentioned ideal situation, and the length oscillations are delayed with respect to the temperature oscillations. For real modulations, a frequency-dependent complex calibration coefficient (correction coefficient) k is necessary to obtain the corrected reversible thermal expansion coefficient _αrev-corr : $$\begin{array}{l}{\mathrm{\alpha }}_{\mathrm{rev}-\mathrm{corr}}=\mathrm{k}\times {\mathrm{\alpha }}_{\mathrm{rev}}\\ \mathrm{.DELTA.L}=\mathrm{k}\times {\mathrm{L}}_{\mathrm{total}}\times {\mathrm{\alpha }}_{\mathrm{rev}}\times {\mathrm{A}}_{\mathrm{T}}\times \sin \left(\mathrm{.omega.t}\right)\mathrm{,}\end{array}$$

If a modulated temperature program (eg the in Fig. 3 shown type) is used, which therefore additionally the underlying, z. B. linear temperature change (see. Fig. 1 ), such a "calibration" can advantageously be carried out so that a calibrated determination of the thermal expansion coefficient α is made possible: $${\alpha }_{\mathit{rev}-\mathit{corr}}=\frac{1}{L_{\mathit{total}}}\times \frac{1}{k}\times \frac{A_L}{A_T}$$

In general, and in particular if the sample is subject to a thermally-induced transformation process at the temperature or temperature change in the course of thermomechanical analysis, have a (directly measured) "total" length change dL _total a "reversible" component dL _rev and a "non-reversible" component dL _nonrev , and the following applies: dL _total = dL _rev + dL _nonrev .

dL _total can be calculated as a mean (eg averaged over exactly one period) of the time-resolved measured value of dL.

wobei dL_{o rev} die reversible Längenänderung am Anfang des Temperatursegmentes ist. Am Anfang der Messung gilt: dL_rev=0.dL _rev can be calculated as ${\mathit{dL}}_{\mathit{rev}}={\mathit{dL}}_{\mathit{o\; rev}}+L_O\times {\int }_{T_0}^T{\alpha }_{\mathit{rev}}\mathit{dT}.$

where dL _{o rev is} the reversible change in length at the beginning of the temperature segment. At the beginning of the measurement: dL _rev = 0.

Accordingly, a _total immediately available when viewed from dL "total" thermal expansion coefficient α has a _total "reversible" component α _rev and a "non-reversible" component _nonrev α, and: α _total = α + α _{nonrev rev.}

In the Fig. 1 to 3 Assuming a purely reversible thermal expansion (dL _total = dL _rev , dL _nonrev = 0, α _total = α _rev , α _nonrev = 0), this means that the sample is not subject to any thermally induced conversion process in the thermomechanical analysis (eg Also, the resulting relative changes in length dULo, where L ₀ denotes an initial length of the analyzed sample (the difference between L and L ₀ is practically negligible in practice).

Assuming that the sample does not undergo a thermally induced transformation process during thermomechanical analysis (α _total = α _rev-corr ), the correction parameter k from formulas (1) and (3) can be calculated as the quotient of the "total expansion coefficient" α _total , which characterizes a "total average change in length" ΔL _avg (considered over a relatively large temperature change ΔT _avg ), and the "reversible Expansion coefficients "α _rev , which characterizes a" reversible component dL _{rev of} the change in length "dL (viewed over a relatively small temperature change dT): $$\mathrm{k}={\mathrm{\alpha }}_{\mathrm{total}}/{\mathrm{\alpha }}_{\mathrm{rev}}$$

The quantities α _total and α _rev required to determine the correction parameter k are available as follows: $${\mathrm{\alpha }}_{\mathrm{total}}=\mathrm{1}/{\mathrm{L}}_{\mathrm{total}}\times \mathrm{d}{\mathrm{L}}_{\mathrm{total}}/\mathrm{d}{\mathrm{T}}_{\mathrm{avg}}\mathrm{and}{\mathrm{\alpha }}_{\mathrm{rev}}=\mathrm{1}/{\mathrm{L}}_{\mathrm{total}}\times {\mathrm{A}}_{\mathrm{L}}/{\mathrm{A}}_{\mathrm{T}}$$

By means of the correction parameter k determined in this way, it is thus advantageously possible to calculate a corrected reversible component dL _{rev-corr of} the change in length dL or a corrected reversible component _{αrev-corr of} the (reversible) thermal expansion coefficient _αrev (compare the above formula 4).

However, this method fails if the sample undergoes a thermally induced transformation process during the modulated temperature program.

It is an object of the present invention, in a TMA of the aforementioned and z. B. in the US 6,007,240 described manner to allow the calculation of a corrected reversible component of the change in length and / or a corrected reversible component of the coefficient of thermal expansion for areas in which the sample undergoes a thermally induced transformation process in the thermomechanical analysis.

This object is achieved in the method according to the invention in that the temperature program has a first segment with a first base heating rate and the correction parameter is calculated on the basis of data from the first segment temperature-dependent from an estimate of a temperature-dependent function of said ratio in a region of the first segment, in which the sample is not subject to a thermally induced conversion process, and that the temperature program, a second segment with a z. B. compared to the first base heating rate smaller second base heating rate and for a calculation of the corrected reversible component the change in length and / or the thermal expansion coefficient in the second segment of the calculated using the data from the first segment correction parameter is used.

The invention is thus based on the idea to calculate the correction parameter including its "temperature dependence" in a suitable (first) segment of the temperature program, and then this calculation result for the determination or calculation of in another (second) segment of the temperature program to be used To use correction parameters. This is possible even if in the second segment the sample is subject to (at least) a thermally induced conversion process.

The use of the temperature-dependent calculated in the first segment correction parameter can, for. B. be provided in the manner of extrapolation.

In one embodiment, it is provided that the amount of the second base heating rate is equal to the amount of the first base heating rate. This can be disadvantageous insofar as - due to the temperature change in the course of the second segment - the use there can bring about a certain error in the result of the correction parameter calculation coming from the first segment.

In another embodiment, it is therefore provided that the absolute value of the second base heating rate is relatively small and / or smaller than the absolute value of the first base heating rate. In extreme cases, even a second heating rate of zero can be provided.

Thus, even against the background that the correction parameter is generally temperature-dependent, the aforementioned falsification can be avoided.

In one embodiment, it is provided that the first base heating rate is greater than zero (corresponding to a modulated heating of the sample in the first segment).

In one embodiment, it is provided that the second base heating rate is greater than or equal to zero (corresponding to a modulated heating or modulated temperature maintenance of the sample in the second segment).

In one embodiment, it is provided that the amount of the first base heating rate is greater than 0.5 K / min, in particular greater than 1 K / min, but this amount z. B. may be less than 100 K / min, in particular less than 50 K / min.

In one embodiment, it is provided that the amount of the second base heating rate is less than 0.5 K / min, in particular less than 0.2 K / min.

In one embodiment, it is provided that the amounts of the first base heating rate and the second base heating rate differ by at least a factor of 2, in particular by at least a factor of 5.

In one embodiment, it is provided that the sample is subject to a thermally induced conversion process within the second segment.

In one embodiment, it is provided that the temperature of the sample rises in the course of the second segment and exceeds a value of 500 ° C, in particular exceeds a value of 1000 ° C. This is advantageous insofar as many materials can only be expected to undergo thermally induced conversion processes at these relatively high temperatures, which, however, can advantageously be investigated by means of the invention by an appropriate choice of the "second segment".

In one embodiment, it is provided that, when evaluating the data, a non-reversible component of the change in length is also determined. Alternatively or additionally, a non-reversible component of the thermal expansion coefficient can also be determined.

In one embodiment, it is provided that a mathematical estimation function is modeled for the correction parameter calculated on the basis of data from the first segment, which indicates the correction parameter as a function of the temperature of the sample, and that for the calculation of the mathematical estimation function corrected reversible component of the change in length, alternatively or additionally, the corrected reversible component of the thermal expansion coefficient, in the second segment to be used correction parameter is calculated. The correction parameter used in the second segment is thus in this case a parameter extrapolated by means of the estimation function from the first to the second segment.

In an embodiment, it is provided that in the calculation of the estimation function a plurality of "interpolation points" (preferably at least 2, in particular at least 5) are used in order to calculate a temperature-dependent estimation function for the correction parameter with relatively high accuracy.

As estimator z. B. a linear function or z. For example, consider a polynomial (eg, parabola).

In one embodiment, the mathematical form of the estimator used in the evaluation of the data is fixed. Alternatively, a user-definable or selectable mathematical form may be used. The "modeling" of the mathematical estimator can be considered in these cases as a calculation of the respective function parameters (eg by the method of least squares or another fitting algorithm).

When passing from a z. B. dynamic (β ≠ 0) first segment to an immediately adjacent second, z. B isothermal (β = 0) segment, or vice versa, advantageously no discontinuity occurs in the temperature-dependent course of the calculated correction parameter (and consequently also not in the course of the calculated reversible variables dL _rev or α _rev ).

In one embodiment, it is provided that, in the course of the temperature program, the second segment is located immediately after the first segment in terms of time, or vice versa.

In a preferred embodiment, a Fourier analysis is used to obtain from the measured data a signal representative of the change in length of the sample Averages, amplitudes and phase shifts (with respect to a temperature signal). A Fourier analysis has z. B. the advantage that influences of a signal noise are not taken into account.

In order to determine mean values, amplitudes and phase shifts, however, alternatively z. B. also considered to accomplish this on the basis of a mean and amplitude determination without Fourier analysis. Averages can be z. B. be calculated as a mathematical mean between immediately adjacent maxima and minima.

A preferred application of the method according to the invention is the so-called dilatometry, a special type of TMA in which, in the narrower sense, the change in length of a sample under negligible force load is determined (eg force of a maximum of 0.5 N). In addition, z. As an application in the context of a DMA (Dynamic Mechanical Analysis) into consideration. The DMA primarily aims to measure a deformation force on a sample as a function of temperature and / or time under defined dynamic mechanical deformation.

• eine Probenaufnahme zum Aufnehmen der Probe,
• eine Temperiereinrichtung zum Temperieren der Probe gemäß eines modulierten Temperaturprogrammes,
• eine Temperaturmesseinrichtung zur Messung einer Temperatur der Probe im Verlauf der Temperierung,
• eine Längenmesseinrichtung zur Messung einer Längenänderung der Probe im Verlauf der Temperierung,
• eine Datenaufzeichnungseinrichtung zum Aufzeichnen von Daten, die repräsentativ für die Längenänderung der Probe im Verlauf der Temperierung sind, und von Daten, die repräsentativ für die Änderung der Temperatur der Probe im Verlauf der Temperierung sind,
• eine Auswerteeinrichtung, die dazu ausgebildet ist, eine reversible Komponente der Längenänderung und/oder eine reversible Komponente des thermischen Ausdehnungskoeffizienten der Probe zu ermitteln und eine korrigierte reversible Komponente der Längenänderung und/oder des thermischen Ausdehnungskoeffizienten mittels eines Korrekturparameters zu berechnen, der unter Verwendung eines aus den Daten ermittelten Verhältnisses eines eine totale Längenänderung charakterisierenden Parameters und eines die reversible Komponente der Längenänderung charakterisierenden Parameters berechnet wird,

dadurch gekennzeichnet, dass die Temperiereinrichtung dazu ausgebildet ist, im Verlauf des Temperaturprogrammes ein erstes Segment mit einer ersten Basisheizrate und ein zweites Segment mit einer (z. B. gegenüber der ersten Basisheizrate kleineren) zweiten Basisheizrate vorzusehen, und dass die Auswerteeinrichtung dazu ausgebildet ist, den Korrekturparameter anhand von Daten aus dem ersten Segment temperaturabhängig aus einer Schätzung einer temperaturabhängigen Funktion des genannten Verhältnisses in einem Bereich des ersten Segmentes zu berechnen, in welchem die Probe keinem thermisch bedingten Umwandlungsprozess unterliegt, und für eine Berechnung der korrigierten reversiblen Komponente der Längenänderung und/oder des thermischen Ausdehnungskoeffizienten im zweiten Segment den anhand der Daten aus dem ersten Segment berechneten Korrekturparameter zu verwenden.According to a further aspect of the invention, there is provided an apparatus for thermomechanical analysis of a sample of material comprising

• a sample receptacle for receiving the sample,
• a tempering device for tempering the sample according to a modulated temperature program,
• a temperature measuring device for measuring a temperature of the sample in the course of tempering,
• a length measuring device for measuring a change in length of the sample in the course of tempering,
• a data recorder for recording data representative of the change in length of the sample during the course of the temperature control and of data representative of the change in the temperature of the sample during the course of the temperature control Tempering are,
• an evaluation device, which is designed to determine a reversible component of the change in length and / or a reversible component of the coefficient of thermal expansion of the sample and to calculate a corrected reversible component of the change in length and / or the thermal expansion coefficient by means of a correction parameter using an off the ratio of a parameter characterizing a total change in length and a parameter characterizing the reversible component of the change in length is calculated,

characterized in that the tempering device is designed to provide a first segment having a first base heating rate and a second segment having a second base heating rate (eg, smaller than the first base heating rate) during the temperature program, and in that the evaluation device is designed to calculate the correction parameter on the basis of data from the first segment as a function of temperature from an estimate of a temperature-dependent function of said ratio in a region of the first segment in which the sample does not undergo a thermally induced transformation process, and for a calculation of the corrected reversible component of the change in length and / or the thermal expansion coefficient in the second segment to use the correction parameter calculated from the data from the first segment.

The particular embodiments and developments described for the analysis method according to the invention can be provided, individually or in any combination, in an analogous manner as special embodiments or developments of the analysis device according to the invention.

According to a further aspect of the invention, a computer program product is provided, comprising a program code which carries out an analysis method of the type described here on a data processing device. Such a computer program product can be used in particular for programming or for sequence control of a control unit contained in the analysis device provided according to the invention.

Fig. 1

ein Diagramm zur Veranschaulichung einer linaren Temperaturänderung und der Messung einer dadurch bewirkten Längenänderung an einer Probe,

Fig. 2

ein Diagramm zur Veranschaulichung einer sinusförmigen Temperaturänderung und der Messung einer dadurch bewirkten Längenänderung an der Probe,

Fig. 3

ein Diagramm zur Veranschaulichung eines modulierten Temperaturprogrammes, zusammengestzt aus einer linaren Temperaturänderung (Fig. 1) und einer Temperaturmodulation (Fig. 2), und der Messung einer dadurch bewirkten Längenänderung an der Probe,

Fig. 4

ein Diagramm zur Veranschaulichung des Ergebnisses einer thermomechanischen Analyse gemäß eines ersten Beispiels (Probe aus Metall),

Fig. 5

ein Diagramm zur Veranschaulichung des Ergebnisses einer thermomechanischen Analyse gemäß eines zweiten Beispiels (Probe aus Keramik),

Fig. 6

ein Diagramm zur Veranschaulichung eines Zwischenergebnisses einer thermomechanischen Analyse gemäß eines dritten Beispiels (Probe aus Keramik),

Fig. 7

ein Diagramm zur Veranschaulichung des Ergebnisses der thermomechanischen Analyse gemäß des dritten Beispiels, und

Fig. 8

eine Vorrichtung zur thermomechanischen Analyse einer Probe gemäß eines Ausführungsbeispiels.

The invention will be further described by means of embodiments with reference to the accompanying drawings. They show:

Fig. 1

a diagram for illustrating a linear temperature change and the measurement of a change in length caused thereby on a sample,

Fig. 2

a diagram for illustrating a sinusoidal temperature change and the measurement of a change in length caused thereby, at the sample,

Fig. 3

a diagram illustrating a modulated temperature program, composed of a linear temperature change ( Fig. 1 ) and a temperature modulation ( Fig. 2 ), and the measurement of a change in length caused thereby,

Fig. 4

3 is a diagram illustrating the result of a thermomechanical analysis according to a first example (metal sample);

Fig. 5

a diagram illustrating the result of a thermomechanical analysis according to a second example (sample of ceramic),

Fig. 6

FIG. 4 is a diagram illustrating an intermediate result of a thermomechanical analysis according to a third example (ceramic sample); FIG.

Fig. 7

a diagram illustrating the result of the thermomechanical analysis according to the third example, and

Fig. 8

a device for thermomechanical analysis of a sample according to an embodiment.

The Fig. 1 to 3 have already been explained above and illustrate a linear temperature change ( Fig. 1 ), a temperature modulation ( Fig. 2 ), and a modulated temperature program resulting from a combination of these temperature changes ( Fig. 3 ).

Fig. 4 illustrates the result of a non-inventive thermomechanical analysis according to a first example, in which the in Fig. 3 shown temperature program was used and the sample analyzed consists of a metal, which is subject to the thermo-mechanical analysis, ie here over a temperature range of about 60-110 ° C, no thermally induced conversion process.

In this respect, this example corresponds to the example already described Fig. 3 , however, in Fig. 4 as a result of the TMA, in particular the time-dependent curves of the corrected quantities dL _rev-corr and α _rev-corr obtained by means of the mentioned correction parameter k are drawn.

Furthermore, in Fig. 4 the curves of the total relative change in length dL _total and the non-reversible relative change in length dL _{nonrev are plotted} . dL _nonrev was obtained here as dL _nonrev = dL _total - dL _rev-corr .

Finally, in Fig. 4 nor the course of the total thermal expansion coefficient α _total drawn.

Since the time-dependent course of the temperature T, ie the "temperature program" is known, can be in accordance with Fig. 4 obtained gradients readily convert into temperature-dependent gradients of the drawn sizes. In the case of the base heating rate β, which is constant here, the in Fig. 4 identifiable processes even qualitatively exactly the respective "temperature-dependent progressions" (cf. Fig. 4 additionally drawn temperature axis). Otherwise, in the case of a non-constant base heating rate, the time-dependent curves _could also be easily "converted" taking into account the time-dependent course of the averaged temperature T _avg .

The determination of the correction parameter k is completely unproblematic in this example, since the sample is not subject to any thermally induced conversion process in the course of the modulated temperature program.

However, the invention aims to calculate a corrected reversible component of the change in length dL _rev-corr, and / or a corrected reversible component coefficient of thermal expansion α _rev-corr , even if the sample undergoes a thermally induced transformation process. An example of this shows Fig. 5 ,

Fig. 5 shows the result of a TMA according to the invention (here: Dilatometrie) according to an example in which a modulated temperature program has been used and the sample analyzed consists of a ceramic, in which by a temperature in the range of about 500-600 ° C, a thermally induced conversion process is expected. In Fig. 5 a section of the TMA result for the temperature range of about 350 ° C to 700 ° C is shown.

A special feature of this example is that the temperature program has a first segment S1 with a first base heating rate β1 (here, for example, about 5 K / min), and the correction parameter k is calculated on the basis of data from this first segment S1, and the temperature program has a second segment S2 with a second base heating rate β2 (here, for example, about 5 K / min), and for a calculation of the corrected reversible component dL _rev-corr the change in length dL in the second segment S2 based on the data from the first segment S1 temperature-dependent calculated correction parameter k is used.

The correction parameter k calculated in the (for lack of thermally induced conversion process) segment S1 of the temperature program calculated for this purpose is to a certain extent extrapolated into the second segment S2 which adjoins to higher temperatures (and there to calculate the gradients of dL _rev-corr and α _rev-corr used).

For example, the value of k obtained as the result of the correction parameter computation performed in the first segment S1 for the end of the first segment S1 (ie at about t = 128 min or T = 460 ° C) than that over the entire second segment S2 for the Correction parameters k to be corrected are to be provided. However, as a result of the correction parameter calculation carried out in the first segment S1, a mathematical function describing the temperature-dependent course of the correction parameter k resulting in the first segment S1 is preferably modeled, in order then to insert the temperatures prevailing in the second segment into it Function to obtain the temperature-dependent correction parameter for correction in the second segment S2 (extrapolation).
This advantageously permits the correction of the reversible variables also in the second segment S2, although there the sample is subject to the thermally induced conversion process.

In contrast to the illustrated example, in which the base heating rates β1, β2 of the segments S1, S2 are selected to be equal (β1 = β2 = 5 K / min), any distortion of the correction extended to the segment S2 by "extrapolation" could be due to a temperature dependence of the correction parameter k can be reduced or avoided by selecting the base heating rate β2 used in the second segment S2 to be comparatively small (or even to β2 = 0).

Im in Fig. 5 1, the correction parameter k was calculated by means of the recorded data from the first segment S1 in the form of a temperature-dependent correction estimation function, for example by means of a known fit (for example according to the least squares method or the like). a predetermined mathematical function which indicates the correction parameter k as a function of the temperature T of the sample. By means of this estimation function, the correction parameter k for use in the second segment S2 was then calculated and used to calculate the gradients of dL _rev-corr and α _rev-corr (and from this the remaining ones in _FIG Fig. 5 shown sizes).

The Fig. 6 and 7 show a further example of a dilatometric analysis of a sample of a ceramic according to the invention.

Fig. 6 shows the modulated temperature program used, which has a first modulated segment S1 in the time domain of about 100 to 280 min and immediately thereafter a second modulated segment S2 in the time domain of about 280 to 400 min.

In the first modulated segment S1, the temperature program has a first base heating rate β1 of about 5 K / min, so that the averaged temperature T _avg increases linearly from about 315 ° C to about 1225 ° C.

In the second segment S2 a much smaller, second base heating rate β2 of about 0.1 K / min is provided so that the average temperature T _avg, starting from about 1225 ° C (final temperature of the first segment S1) linearly only by about 10 ° C. rises to about 1235 ° C.

During the course of the temperature program, the sample investigated here shows several thermally induced transformation processes, namely at temperatures of around 500 ° C, 945 ° C and 1130 ° C. Thus, while the first two "effects" (at 500 ° C and 945 ° C) are in the first segment S1, the latter effect (at 1130 ° C) extends beyond the boundary between the two segments S1, S2. The effect starts at about 1130 ° C and does not end at 1225 ° C.

(For the calculation of the corrected reversible component dL _rev-corr see Fig. 7 ) of the immediately measured total length change dL _total ( Fig. 6 ) a correction parameter k was calculated as a function of temperature on the basis of data from the first segment S1, namely in a range of S1 in which no thermal effect occurs, that is, for. B. based on data from the time range of about 100 to 130 minutes, corresponding to a temperature range of about 315 to 470 ° C.

The correction parameter k was calculated here again in the form of an estimate of the temperature-dependent function of the ratio of the total thermal expansion coefficient α _total and the reversible coefficient of thermal expansion α _rev , these coefficients of expansion being in the range of [100 min, 130 min] and [315 ° C , 470 ° C] directly measured or known data (T _avg , ΔT, L, dL) were obtained. As estimation z. B. in the simplest case a linear function can be used (change of k proportional to change in temperature). Deviating from this, however, could also be complicated estimators, z. As polynomials at least 2nd degree, can be used.

However, the correction parameter k calculated in this way as a temperature-dependent function is advantageously not only used for the intended correction in the range [100 min, 130 min] or [315 ° C, 470 ° C], but (by extrapolation) over the entire modulated one Temperature program, ie [100 min, 400 min] or [315 ° C, 1235 ° C]. Notwithstanding the example described ( Fig. 6 and 7 ), in which a small but non-vanishing base heating rate β2 of approximately 0.1 K / min is provided in the segment S2, the base heating rate β2 used in the second segment S2 could also be zero. Very advantageous may be a very small second base heating rate or β2 = 0, for example, if at the beginning of the second segment, the temperature is so high that a further increase in the temperature in the course of the second segment S2 prepares equipment difficulties.

Also different from the example described ( Fig. 6 and 7 In the context of the invention, the second segment S2 (with low heating rate β2) could also precede the first segment S1 (with high heating rate β1) in which the second segment S2 follows the first segment S1 in terms of time. This would be z. B. useful if the sample to be examined is subject to one or more thermally induced transformation processes at rather low temperatures, whereas at rather high temperatures no such conversion processes take place. Even then, the correction parameter k calculated in the suitable segment (which contains a region without a thermally induced transformation process) of the temperature program can be extrapolated into the temporally preceding (at lower temperatures) second segment S2, where it is used to calculate the gradients of dL _{revolutions. corr} and / or α _rev-corr .

According to the analysis methods described above, an analysis for dynamic segments (heating rate β1 greater or smaller than 0) and isothermal segments (β2 = 0) or nearly isothermal segments (β2 <β1 or β2 << β1) is advantageously possible simultaneously.

The temperature-modulated measurements provided according to the invention can advantageously be extended into a relatively high temperature range, e.g. B. up to operationally approached by the temperature program end temperatures of T _avg greater than 1000 ° C, in particular greater than 2000 ° C, so that a possible application z. B. is the analysis or characterization of the sintering behavior of ceramics.

A calibration of the measurement of reversible material parameters (such as dL _rev and α _rev ) performed for a dynamic segment (β ≠ 0) of the temperature program by means of a correction parameter k can be advantageous for one or more other segments, in particular z. B. for an immediately preceding or an immediately following segment (in particular, for example, with a relatively small or even as 0 selected β) are taken into account in the calculation of the analysis result.

Fig. 8 1 shows an embodiment of an apparatus 10 for carrying out an analysis method of the type described here. The apparatus 10 is used in this example in the context of a thermomechanical analysis (TMA) (Alternatively, the apparatus 10 could also be in a DMA (Dynamic Mechanical Analysis) mode, So with variable power preload, operated).

Part of the device components are in Fig. 8 shown in dashed lines, namely a sample holder 1 in a controlled temperature oven 3 for storage and controllable temperature control of a sample P, at which the TMA measurement is performed.

During the measurement, the change in length (in this case, for example, resulting from a temperature-modulated heating or cooling of the sample P in the furnace 3) is measured in a direction which is referred to below as the longitudinal direction and in FIG Fig. 8 indicated by the double arrow A. In addition, in the measurement, the sample P is subjected to a predetermined constant small force load (eg, about 0.1 N), which also acts on the sample P in the longitudinal direction A.

During the length (change) measurement, the sample P received by the sample holder 1 is tempered according to a modulated temperature program (eg of the type already described) and the force exerted on the sample is regulated to a predetermined constant value.

The device 10 comprises a feeler punch 12, which exerts on the sample P the predetermined force (here: pressure force) during the measurement with a feeler end 14.

The material of the Fühlstempels should have the lowest possible or at least well-defined thermal expansion in this application, so that this thermal expansion affects the measurement result little or can be considered accordingly within a control program and / or a subsequent evaluation. The same applies to the material of the feeler recording.

• eine stationäre Basis 22, die im dargestellten Beispiel ein erstes Basisteil 22-1 (z. B. zum Haltern des Ofens 3 bezüglich der Messeinrichtung 20) und ein zweites, langgestreckt in Längsrichtung A sich erstreckendes Basisteil 22-2 umfasst,
• eine an der stationären Basis 22 über eine Führungseinrichtung 24 in Längsrichtung A relativ zur stationären Basis 22 bewegbar gelagerte Fühlstempelaufnahme 26, an welcher der Fühlstempel 12 im dargestellten Beispiel mittels einer Schraubklemme 28 befestigt ist,
• eine ansteuerbare Antriebseinrichtung 30 zum Antreiben der Fühlstempelaufnahme 26 in Längsrichtung A relativ zur stationären Basis 22,
• eine Kraftmesseinrichtung 32 zum Erfassen der vom Fühlstempel 12 auf die Probe P ausgeübten Kraft,
• eine Steuereinrichtung ST, die dazu ausgebildet ist, die Antriebseinrichtung 30 abhängig von der mittels der Kraftmesseinrichtung 32 erfassten Kraft gemäß der vorbestimmten Kraftbelastung anzusteuern, und
• einen Wegaufnehmer 34 zur Messung der Bewegung des Fühlstempels 12 in Längsrichtung A relativ zur Basis 22.

The device 10 further comprises a measuring device 20 to which the feeler 12 is attached and by means of which the movement of the feeler 12 resulting from the change in length of the sample P in the longitudinal direction A is measured during the measurement. The measuring device 20 comprises:

• a stationary base 22, which in the illustrated example comprises a first base part 22-1 (eg for holding the furnace 3 with respect to the measuring device 20) and a second, elongated longitudinal part A extending base part 22-2,
• a sensing-stamp receiving means 26 movably mounted on the stationary base 22 via a guide means 24 in the longitudinal direction A relative to the stationary base 22, to which the sensing punch 12 is fastened by means of a screw clamp 28 in the example shown,
• a controllable drive device 30 for driving the feeler die receptacle 26 in the longitudinal direction A relative to the stationary base 22,
• a force measuring device 32 for detecting the force exerted on the sample P by the feeler 12,
• a control device ST, which is designed to control the drive device 30 as a function of the force detected by the force measuring device 32 in accordance with the predetermined force load, and
• a displacement sensor 34 for measuring the movement of the feeler plunger 12 in the longitudinal direction A relative to the base 22.

• Ausgehend von der in Fig. 8 dargestellten Situation, bei welcher die Probe P bereits am Probenhalter 1 im Ofen 3 angeordnet ist und am dem Probenhalter 1 abgewandten Ende vom Fühlstempelende 14 kontaktiert wird, wird mittels eines in der Steuereinrichtung ST ablaufenden Steuerprogrammes ein "TMA-Messprogramm" durchgeführt, welches darin besteht, die Probe P gemäß des Temperaturprogrammes aufzuheizen, die Probe P hierbei mittels des Fühlstempels 12 gemäß der Kraftvorgabe (hier: zeitunabhängige Druckkraft) vorzubelasten und die sich im zeitlichen Verlauf daraus ergebende Längenänderung der Probe P in Längsrichtung A zu messen.

The function of the illustrated device 10 can be described as follows:

• Starting from the in Fig. 8 the situation in which the sample P is already arranged on the sample holder 1 in the oven 3 and contacted at the end remote from the probe holder end 14, a "TMA measuring program" is performed by means of a running in the control device ST control program, which therein In this case, the sample P is heated according to the temperature program, the sample P is preloaded by means of the feeler 12 in accordance with the force specification (here: time-independent compressive force) and the length change of the sample P in the longitudinal direction A is measured.

For this purpose, the control device ST is connected to corresponding device components via control lines L1, L2, L3 and L4. The line L1 serves to transmit a temperature control signal to the furnace 3 or a tempering device integrated therein (eg electric heater). In addition, via the line L1, a retransmission of a z. B. by means of a thermocouple on the sample P measured temperature signal to the controller ST done. The line L2 serves to transmit a force-measuring signal (measured value of the force exerted on the sample P) from the force-measuring device 32 to the control device ST. The line L3 is used to transmit a drive control signal to the drive means 30. The line L4 is used to transmit an output from the position sensor 34 Wegmesssignals.

In particular, the z. B. transmitted over the line L1 time-dependent temperature signal (representative of the temperature T) as well as the transmitted over the line L4 time-dependent length change signal (representative of z., For the absolute change in length dL) is stored in the control device ST to a corresponding evaluation after completion of the measurement.

During the measurement, the control device ST controls the force exerted on the sample P by means of the feeler plunger 12 by means of a control of the drive device 30. This works as follows: As already mentioned, the feeler temple 12 is attached to the feeler stamp base 26 by means of the screw clamp 28, so that a corresponding loading or adjustment of the feeler temple receptacle 26 is transmitted directly to the feeler temple 12 and thus further to the sample P.

In the example shown, the feeler stamp receiver 26 is not driven directly by the drive device 30. Rather, the Fühlstempelaufnahme 26 as in Fig. 8 can be seen connected to a in the figure upper end of an elastic body 32-1 (here: screwed), whose in Fig. 8 lower end connected to a carriage 24-1 of the guide means 24 (here: screwed), which in the longitudinal direction A moved to a guide rail 24-2 guided the guide means 24 and by means of a push rod 30-1 of the drive means 30 is moved accordingly. The end of the push rod 30-1 is screwed to the slide 24-1 as seen in the figure. Accordingly, an activation of the drive device 30, which in the example illustrated contains a piezo stepping motor controlled via the line L3, initially causes a displacement of the push rod 30-1 and thus of the carriage 24-1 in the longitudinal direction A. Via the elastic body 32-1 this becomes Carriage movement via the Fühlstempelaufnahme 26 and the feeler 12 converted into a corresponding force or force change to the sample P.

Thus, the unit formed from the carriage 24-1, the feeler stem receiver 26, and the elastic body 32-1 interposed therebetween may also be referred to as a drivable "feeler stamp base" attached to the stationary base 22 via the guide means 24 in the longitudinal direction A relative to the stationary base 22 is movably mounted and on which the feeler 12 is attached (or fastened). The feeler stamp base 24-1, 26, 32-1 is driven in the longitudinal direction A relative to the stationary base 22 as explained by the controllable drive device 30.

The elastic body 32-1 forms the force measuring device 32 together with a strain gauge (not shown in the figure) connected to the line L2 (for example, attached to an outer surface of the elastic body 32-1).

In particular, when the elastic body 32-1 is formed of a metallic material, so instead of a solid material, a z. B. be used at least in a central area recessed material, as in Fig. 8 dashed symbolizes (here: circular recess in a cuboid body). Deviating from z. B. also be provided in each case a plurality of smaller such recesses of the elastic body.

By means of the force measuring device 32, the actual force exerted on the sample P force is measured as a function of time during the measurement and preferably as an analog / digital converted signal via the line L2 of the control device ST fed (or in the control device analog / digital converted) ,

In the controller ST is also the z. B. by a corresponding user input predetermined force load, d. H. the force to be applied to the sample P ("target value") is stored. By means of the running in the control device ST control program, the measured "actual value" of the force is controlled by appropriate control of the drive means 30 to the predetermined by the predetermined force load "target value". Advantageously, a friction which is unavoidable in practice in the region of the guide device 24 (between the slide 24-1 and the guide rail 24-2) does not lead to a corresponding falsification of the force measurement.

Likewise, the simultaneous measurement carried out the change in length of the sample P in the illustrated embodiment does not lead to a distortion of the force measurement. In the example shown, the measurement of the change in length is realized as follows: A scale carrier 34-1 with a trained therefrom or in the example shown attached thereto (for example, the movement in the longitudinal direction A corresponds to the change in length of the sample P to be measured Glued scale) 34-2 attached so that the change in length of the sample P transmits in a corresponding displacement of the scale 34-2, which can thus be measured in a simple manner.

For this purpose, the displacement transducer 34 further comprises a stationary held optical sensor 34-3 for measuring the relative displacement between this sensor 34-3 and the scale 34-2. The scale 34-2 is z. B. provided with equidistant over its length distributed bar marks or formed, which are detected when passing the scale 34-2 by means of the optical sensor 34-3 and connected thereto (or integrated therein) linear encoder electronics to a for the change in length of the sample P. to obtain a representative measurement signal (the number of detected markings is a measure of the change in length). This measurement signal is supplied via the line L4 to the control device ST and stored there digitally for later evaluation as a time-dependent sample length signal.

The linear encoder electronics can z. B. in the region of the optical sensor 34-3 or alternatively be arranged in the region of the control device ST.

The stationary holding of the optical sensor 34-3 is accomplished in the example shown by a holder 34-4, which on the one hand connected to the sensor 34-3 (here: screwed) and on the other hand connected to the base 22 (here: screwed).

By means of the illustrated device 10 can thus be advantageously measured in a TMA analysis, the change in length of the predetermined force in the longitudinal direction A sample P, wherein the resulting from the change in length of the sample P in the longitudinal direction A movement of the feeler 12 is measured without contact, wherein driving the feeler stamp base 24 - 1, 26, 32 - 1, to which the feeler punch 12 is attached, takes place in the longitudinal direction A relative to the stationary base 22, and this driving by means of the control software depends on the pressure exerted on the sample P by the feeler 12 Force detected by the force measuring device 32 according to the predetermined force load takes place (force control).

By means of the software running in the control device ST, the recorded data are subjected to an evaluation in order to determine temperature-dependent profiles of the reversible and non-reversible components of the change in length dL (or dL / L) or of the thermal expansion coefficient α. By appropriate design of this software, in particular, the embodiments and developments described above for the analysis method can be realized.

Claims (12)

A method of thermomechanical analysis of a sample (P) of a material comprising a) arranging the sample (P) in a thermomechanical analysis device (10) and tempering the sample (P) by means of the analysis device according to a modulated temperature program, b) recording data obtained by means of the analysis device, which are representative of a change in length (dL) of the sample (P) in the course of the temperature control, c) evaluating the data to determine a reversible component (dL _rev ) of the change in length (dL) and / or a reversible component (α _rev ) of the thermal expansion coefficient (α) of the sample (P), d) calculating a corrected reversible component (dL _rev-corr ; a _rev-corr ) of the change in length (dL) and / or the coefficient of thermal expansion (α) by means of a correction parameter (k) using a ratio determined from the data of a total length change (dL _total ) characterizing parameter (α _total ) and a parameter (α _rev ) characterizing the reversible component (dL _rev ) of the change in length (dL) is calculated, characterized in that the temperature program has a first segment (S1) with a first base heating rate (β1) and the correction parameter (k) based on data from the first segment (S1) temperature dependent on an estimate of a temperature dependent function of said ratio in a range of first segment (S1) is calculated, in which the sample (P) is not subject to a thermally induced conversion process,
and in that the temperature program has a second segment (S2) with a second base heating rate (β2) and for a calculation of the corrected reversible component (dL _rev-corr ; α _rev-corr ) the change in length (dL) and / or the thermal expansion coefficient (α ) in the second segment (S2) the correction parameter (k) calculated on the basis of the data from the first segment (S1) is used. The method of claim 1, wherein the amount of the second base heating rate (β2) is equal to the amount of the first base heating rate (β1). The method of claim 1, wherein the amount of the second base heating rate (β2) is less than the amount of the first base heating rate (β1). Method according to one of the preceding claims, wherein the amount of the first base heating rate (β1) is greater than 0.5 K / min, in particular greater than 1 K / min. Method according to one of the preceding claims, wherein the amount of the second base heating rate (β2) is less than 0.5 K / min, in particular less than 0.2 K / min. Method according to one of the preceding claims, wherein the amounts of the first base heating rate (β1) and the second base heating rate (β2) differ from each other by at least a factor of 2, in particular by at least a factor of 5. Method according to one of the preceding claims, wherein the sample (P) within the second segment (S2) undergoes a thermally induced transformation process. Method according to one of the preceding claims, wherein the temperature (T) of the sample (P) in the course of the second segment (S2) increases and exceeds a value of 500 ° C, in particular exceeds a value of 1000 ° C. Method according to one of the preceding claims, further comprising a non-reversible component when evaluating the data (dL _nonrev) the change in length (dL) is determined. Method according to one of the preceding claims, wherein for the correction parameter (k) calculated on the basis of data from the first segment (S1) a mathematical estimation function is modeled which determines the correction parameter (k) as a function of the temperature (T) of the sample (P). and using the mathematical estimation function to calculate the correction parameter (k) to be used for calculating the corrected reversible component (dL _rev-corr ) of the change in length (dL) in the second segment (S2). Method according to one of the preceding claims, wherein in the course of the temperature program, the second segment (S2) in terms of time is immediately after the first segment (S1), or vice versa. Apparatus (10) for thermomechanical analysis of a sample (P) of a material comprising a sample receptacle (1) for receiving the sample (P), a tempering device (ST, L1, 3) for tempering the sample (P) according to a modulated temperature program, a temperature measuring device (L1) for measuring a temperature (T) of the sample (P) in the course of the temperature control, a length measuring device (20) for measuring a change in length (dL) of the sample (P) in the course of the temperature control, a data recorder (ST) for recording data representative of the change in length (dL) of the sample (P) during the course of the temperature control and data representative of the change in temperature (dT) of the sample (P) in the Course of tempering are, - An evaluation device (ST), which is adapted to a reversible component (dL _rev ) of the change in length (dL) and / or a reversible Component (α _rev ) of the coefficient of thermal expansion (α) of the sample (P) to determine and a corrected reversible component (dL _{rev -corr} ; a _rev-corr ) of the change in length (dL) and / or the coefficient of thermal expansion (α) by means of a Correction parameter (k) calculated using a ratio of a total length change (dL _total ) characterizing parameter (α _total ) and a parameter (α _rev ) characterizing the reversible component (dL _rev ) of the change in length (dL). is calculated, characterized in that the tempering device (ST, L1, 3) is designed to provide in the course of the temperature program a first segment (S1) with a first base heating rate (β1) and a second segment (S2) with a second base heating rate (β2),
and in that the evaluation device (ST) is designed to calculate the correction parameter (k) on the basis of data from the first segment (S1) as a function of temperature from an estimate of a temperature-dependent function of said ratio in a region of the first segment (S1) in which the sample (P) is not subject to a thermally induced transformation process, and for a calculation of the corrected reversible component (dL _rev-corr ; α _rev-corr ) the change in length (dL) and / or the thermal expansion coefficient (α) in the second segment (S2) to use the correction parameter (k) calculated from the data from the first segment (S1).

Images